OK – I did a test of this idea, and I have good news and bad news.
Good news
Making a tea of salvia leaves and collecting the sediment yields salvinorin A!
Bad news
The yield is about half of what I’d get if using acetone as the initial solvent.
----------
Details
I started with 30g of dried salvia leaf (very fresh – harvested just a few weeks ago). I made a tea using about 300ml of water. I brought the water to a boil with the leaves, then removed it from heat. I let the tea steep for about 10 minutes (same as I’d do with any tea).
I then stirred the mix to make sure the salvinorin (if any) was suspended, and poured the tea out through a sieve.
I did this two more times, ending up with about 1 liter of tea. I didn’t reduce it – I was concerned about what might happen to salvinorin A when heated in aqueous suspension (in crystal form, I think it’s very stable, but I didn’t want to risk destroying it).
I let the tea sit overnight, and noticed that sediment had collected: it was a mix of some leaf particles that passed through the sieve and a finer sediment. I decanted most of the water, then put the sediment plus the remaining water in a smaller container and let it settle again.
When it was settled, I decanted the remaining tea (as much as I could without disturbing the sediment) and then added 50ml acetone to the sediment. I stirred, then allowed it to settle.
After settling, I decanted the acetone and discarded the sediment, which was mostly leaf particles.
I put the acetone in a shallow pan and allowed it to evaporate and the residue to dry thoroughly. I scraped up the residue, and it yielded a light green, sparkly powder – looked just like crude salvinorin A. – That’s the good news.
Now the bad news – I weighed the powder: 68mg.
If this batch of salvia is about as potent as last years (about 2.7mg/g), I should have gotten a yield of over 81mg of pure salvinorin A. The quantity of powder I have is too small to further refine, but my guess is that I got about 40mg salvinorin A. So the bad news is that the yield isn’t nearly as high as it would be when using acetone as the initial solvent.